PROJECT SUMMARY Our group has demonstrated that optical coherence tomography angiography (OCTA) can address many of the disadvantages of fluorescein angiography (FA) in diabetic retinopathy (DR). OCTA uses the intrinsic contrast of blood cell motion and therefore does not require dye injection, making it entirely non-invasive and suitable for routine examination. OCTA is three-dimensional and can detect retinal neovascularization by the presence of flow in the normally avascular vitreous space, and can detect capillary dropout in the retinal and choroidal circulations separately. The OCTA technology used by previous studies is limited by the small fields of view (2.4-6 mm) and the projection artifacts in the deeper layers, which are caused by flowing blood cells in the more superficial vessels. The current proposal will overcome the previous limitations by improving both the hardware platform and software algorithms to obtain wide-field (WF) and projection-resolved (PR) OCTA. 1. Develop wide-field (WF) OCT system for imaging peripheral retinal circulation. We have developed a 200-kHz swept-source optical coherence tomography (SS-OCT) system. This system is ready for use in clinical studies. We propose to further increase system speed and develop automatic registration and montage algorithms to create ultrawide-field OCTA. 2. Improve the projection-resolved (PR) OCTA algorithm for imaging of retinal and choroidal plexuses. We have preliminarily demonstrated that separation of three retinal plexuses improved the detection of early vascular changes in DR. We will refine this algorithm and apply it to the WF-OCTA on the high-speed SS-OCT systems. 3. Quantify neovascularization and nonperfusion areas using WF- and PR-OCTA in DR. We will further optimize the automated quantification of all relevant DR endpoints, including nonperfusion of all vascular layers, neovascularization, cyst volume, and retinal thickness maps on both commercial and custom OCT systems proposed in this study. 4. Evaluate advanced OCTA for DR in clinical studies. The clinical studies with DR and control subjects are as follows: (1) cross-sectional study to determine if WF- and PR-OCTA can detect nonperfusion and other microvasculopathies with better sensitivity than conventional OCTA and if they can detect these changes even in diabetics without photographic signs of DR; (2) long-term prospective observational study to determine if quantitative WF- and PR-OCTA parameters can predict the risk of DR progression; and (3) short-term study in patients undergoing treatment for diabetic macular edema or neovascularization to validate WF- and PR-OCTA measurements of nonperfusion and neovascularization in comparison with ultrawide-field FA.